-A189 EohhEEEohEoiE EhhEEEEEEElhhE EEmhhhmhhmhhhE E E … · -A189 019 BNF ADVANCED AVIONICS T SYS...
Transcript of -A189 EohhEEEohEoiE EhhEEEEEEElhhE EEmhhhmhhmhhhE E E … · -A189 019 BNF ADVANCED AVIONICS T SYS...
-A189 019 ADVANCED AVIONICS SYS BNF T ESUYMO ( ANALYI SCDLIIEONC S CO ST IRBOO
ON Id L BEDZYK ET AL. FEB 87 TASC-J-5843
UNCLASSIFIED AFWAL-_71-87-H18 F33615-83-C 1053 F/6 9/1i U
EohhEEEohEoiEEhhEEEEEEElhhEEEmhhhmhhmhhhEE E Em o h m h h h o h h h I
EEEEEsohhEEE ohE
• IIII1
MICROCOPY RESOLUTION TEST CHART
0 | • •A• 1 . A' Q 8 2.5
05G
1,
@1 W U U U U U U U U2.2
AD-A189 019
AFW'AL-TR-87- 1138
ADVANCED AVIONICS SYSTEM ANALYSIS
Modular Avionics Cost Benefit Study Formulation .
The Analytic Sciences Corporation3040 Presidential DriveFairborn OH 45324
February 1987
Final Report for Period September 1983 - January 1986
Approved for public release; distribution is unlimited
DTICE LECTE
FEB 0 1988-
AVIONICS LABORATORYAIR FORCE WRIGHT AERONAUTICAL LABORATORIESAIP FORCE SYSTEMS COMMANDWRIGHT-PATTERSON AIR FORCE BASE, OHIO 45433-6543
88 2 05 113~ I
NOTICE
When Government drawings, specifications, or other data are used for anypurpose other than in connection with a definitely Government-relatedprocurement, the United States Government incurs no responsibility or anyobligation whatsoever. The fact that the Government may have formulated or inany way supplied the said drawinos, specifications, or other data, is not tobe regarded by implication, or otherwise in any manner construed, as licersirothe holder, or any other person or corporation; or as conveying any rights orperrrission to manufacture, use, or sell any patented invention that may in anyway be related thereto.
This report has been reviewed by the Office of Public Affairs (ASD/PA)and is releasable to the National Technical Information Service (NTIS). AtNTIS, it will be available to the general public, including foreign nations.
This technical report has been reviewed and is approved for publication.
KA .PURDY, lb 1 SAF H,4N C. OSTGAARD, 151efProject Engineer /Advanced Integration GroupAdvanced Integration Group System Integration Branch
FOP THE COM!IANDER
CHARLES H. KRUEGER, JR
Acting Chief
System Avionics DivisionAvionics Laboratory
e "If your address has changed, if you wish to be removed fro our mailinqlist, or if the addressee is no longer employed by your organization pleasenotify AFWAL/AAAS-1, Wright-Patterson AFB, OH 45423-F543 to help us mairtaira current mailing list.
Copies of this report should not be returned unless return is required bv
security considerations, contractual obligations, or notice on a specificdocument.
UNCLASSI FIED
SECURITY CLASSIFICATION OF THIS PAGE'~ Frm ApprovedREPORT DOCUMENTATION PAGE o BNo.0-01
la. REPORT SECURITY CLASSIFICATION Ib. RESTRICTIVE MARKINGS
UNCLASSIFIED
2a. SECURITY CLASSIFICATION AUTHORITY 3 DISTRIBUTION /AVAILABILITY OF REPORT
,2b. DApproved for public release;
2b. ECLSSIFCATON DOWGRAING CHEULEdistribution is unliiited
4. PERFORMING ORGANIZATION REPORT NUMBER(S) 5 MONITORING ORGANIZATION REPORT NUMBER(S)
TASC No J-5043 AFWAL-TR-87-1138
6a. NAME OF PERFORMING ORGANIZATION 6b. OFFICE SYMBOL 7a. NAME OF MONITORING ORGANIZATION(If applicable) Air Force Wright Aeronautical Laboratories
The Analytic Sciences Corp Avionics Laboratory (AFWAL/AAAS-1)6c. ADDRESS (City, State, and ZIP Code) 7b. ADDRESS (City, State, and ZIP Code)
3040 Presidential DriveFairborn OH 45324 Wright-Patterson Air Force Base, Ohio 45433-
65438a. NAME OF FUNDING/SPONSORING 8b. OFFICE SYMBOL 9. PROCUREMENT INSTRUMENT IDENTIFICATION NUMBER
ORGANIZATION (If applicable)F33615-83-C-1053
8c. ADDRESS (City, State, and ZIP Code) 10 SOURCE OF FUNDING NUMBERSPROGRAM PROJECT TASK WORK UNITELEMENT NO. NO. NO ACCESSION NO.
63253F 2734 01 25
11. TITLE (Include Security Classification)Advanced Avionics System AnalysisModular Avionics Cost Benefit Study Formulation
12. PERSONAL AUTHOR(S)
Bedzyk, William L., Czech, Donald R., Dickman, Thomas J., Gruber, Frank S., Myers, John F.
13a. TYPE OF REPORT 13b TIME COVERED 14. DATE OF REPORT (Year, Month, Day) 15. PAGE COUNT
Final FROM 09-83 TO 01-86 February 1987 6916. SUPPLEMENTARY NOTATION
17. COSATI CODES 18. SUBJECT TERMS (Continue on reverse if necessary and identify by block number)
FIELD GROUP SUB-GROUP Avionics VHSIC IMIS0 3 u - Cost Analysis TSIMD
LRM MATE19. ABSTRACT (Continue on reverse if necessary and identify by block number)
This is a technical baseline for a cost benefit analysis of optional features of an
advanced modular avionics architecture for the wid-1990s. It provides an outline of a
projected life cycle cost for alternate configurations for future avionics (including
cost for implementation and supportability).
20 DISTRIBUTION/AVAILABILITY OF ABSTRACT 21 ABSTRACT SECURITY CLASSIFICATION
In UNCLASSIFIEDUNLIMITED 0 SAME AS RPT OTI( USERS UNCLASSIFIFfn22a NAME OF RESPONSIBLE INDIVIDUAL 22b TELEPHONE (Include Area Code) 22c OFFICE SYMBOLKARL S. PURDY, ILt (513) 255-47 09 AFWAL/AAAS-1
D Form 1473, JUN 86 Previous editions are obsolete SECURITY CLASSIFICATION OF THIS PAGE
UNCLASSIFIED
%
FOREWORD
This Technical Report has been written todocument the initial formulation of a technicalbaseline for a cost benefit analysis of optionalfeatures of an advanced modular avionics architec-ture for the mid-1990s. The advantages offered bysuch features as line replaceable modules usingVHSIC and software for dynamic reconfiguration havebeen assumed but little attention has been paid to
*i interactions among the factors or projected costs ofincorporating them into a system. Selection offeatures from a set of alternatives must be made onthe basis of total life cycle cost impact and thebenefit to the overall weapon system performance.Many of the factors set out to improve mission com-pletion reliability. Not all of them are equally
* .efficient in improving performance for the costexpended. This study includes discussion of thefactors and establishes the baseline for a lifecycle cost analysis of alternatives. Estimates oflife cycle cost impacts coupled with the relativebenefits to be gained will allow the decision makerto select the most effective strategy for imple-menting modular avionics.
0
Acoession For
NTIS GRAIV"DTIC TAB El
0 @,. UnannocuodJust v
Di~~Lstrli"'*!-
iii _ _ _ _
TABLE OF CONTENTS
Page
1. INTRODUCTION 1-1
2. BACKGROUND 2-1
3. STUDY DESCRIPTION 3-13.1 Program Management 3-3
3.1.1 Certification Facility 3-33.1.2 Configuration Management Library 3-3
3.2 Acquisition and Development 3-43.2.1 Product Performance Agreements 3-43.2.2 Competition 3-5
3.3 Design 3-63.3.1 Partitioning 3-63.3.2 Standardization 3-83.3.3 Testability 3-93.3.4 Time Stress Measurement Device (TSMD) 3-93.3.5 Reconfiguration/Software
Development Needs 3-103.4 Operation 3-113.5 Support 3-11
3.5.1 Modular Automatic Test Equipment (MATE) 3-11* 3.5.2 Integrated Maintenance Information
System (IMIS) 3-123.5.3 Depot Repair Facility 3-123.5.4 Maintenance Strategy 3-13
3.6 Study Organization 3-13
4. DECISION CRITERIA 4-14.1 Program Management 4-1
4.1.1 Certification Facility 4-14.1.2 Configuration Management Library 4-3
4.2 Development and Acquisition 4-5
4.2.1 Product Performance Agreements 4-54.2.2 Competition 4-7
0 V
TABLE OF CONTENTS (Continued)
Page
4.3 Design 4-84.3.1 Partitioning 4-84.3.2 Standardization 4-124.3.3 Testability 4-154.3.4 Time Stress Measurement Device (TSMD) 4-194.3.5 Software Development for a Dynamically
Reconfigurable Architecture 4-224.4 Support 4-23
4.4.1 Modular Automatic Test Equipment (MATE) 4-234.4.2. Integrated Maintenance Information
System 4-234.4.3 Depot Repair Facility 4-234.4.4 Maintenance Strategy 4-25
5. SCENARIO DEVELOPMENT 5-15.1 Avionics Architecture Baseline 5-25.2 Element Description 5-75.3 Operations and Logistics Scenario 5-7
6. Analysis Methodology 6-1
7. Conclusions 7-1
REFERENCES R- 1
BIBLIOGRAPHY B-i
IV
IW
LIST OF FIGURES
Figure Page
1-1 Standard Modular Architecture Cost/BenefitStudy 1-2
3.3-1 Evaluation Criteria for Determining MaintenanceLevel 3-7
3.6-1 Modular Avionics Decision Tree 3-14
4.1-1 Certification Facility 4-1
4.1-2 Configuration Management Library 4-4
4.2-1 Product Performance Agreements 4-5
4.2-2 Competition 4-7
4.3-1 Partitioning 4-8
4.3-2 Standardization Decision Trees 4-12
4.3-3 Testability 4-16
4.3-5 Reconfiguration Software 4-22
4.4-1 Depot Repair Facility 4-24
4.4-2 Maintenance Strategy 4-25
5.1-1 Modular Avionics Architecture 5-3
5.2-1 Number of Modules Required Versus Module Size 5-8
5.3-1 Modular Avionics Acquisition Schedule Baseline 5-9
5.3-2 Simplified 3-Level Maintenance Model 5-10
5.3-3 Simplified 2-Level Maintenance Model 5-11
6-1 Life Cycle Cost Equation 6-2
6-2 Modular Avionics Life Cycle Flow 6-3
6-3 Cost Estimate Process: Flow Organization for
System Trades 6-6
fil7-1 Life Cycle Cost Trade Table 7-3
Vii -
LIST OF TABLES
Table Page
3-1 Factors Influencing a Modular AvionicsArchitecture 3-2
4.3-1 Partitioning Design Criteria 4-9
4.3-2 Standardization Decision Criteria 4-14
4.3-3 Testability Decision Criteria 4-17
5.1-1 Baseline Common Module Set 5-4
5.1-2 Avionics Tasks 5-5
5.1-3 Required Avionics Module Complement 5-6
Viii
.6
~w -z ~ - Wnp rrU rVK'V-1RP lr :V ~WVU - -- ---- -. '
"-, 1. INTRODUCTION
The objective of this study is to set the baseline
for a cost/benefit analysis of alternatives for a modular
avionics architecture for the mid-1990s. Figure 1-1 shows
each phase of the overall cost/benefit analysis. The objec-
tive of each of the alternatives is to improve the mission
performance or reduce the life cycle cost of the avionics
suite. The alternatives have been selected from the many
studies and reports included in the bibliography that have
- -- iaddressed the implementation of modular avionics to improve
avionics performance and reduce cost. Performance, as used
here, is the ability to perform the mission. Both performance
and cost are influenced by reliability. Hardware reliability
influences operation and support costs as well as mission
reliability. If hardware reliability cannot be improved, the
objectives are to improve mission reliability by replicating
the hardware paths (redundancy) and to reduce the costs of
detecting, isolating, and repairing hardware faults that do
occur. The alternatives discussed in this study address one
or both of the objectives. Their efficiency in meeting both
objectives is the fundamental criterion the alternatives are
measured against.
It is necessary to define an operating scenario in
?. order to evaluate life cycle costs using simulation models.
This scenario includes the weapon system and the suite of
avionics on board. A generic fighter aircraft similar to the
F-15 and the F-16 was assumed for the aseline platform. Theavionics suite includes equipment now under development in ASD
lot
"- " -z I.
I~EEn
u4-4
cnh0 " L 0
*L LULJ 1->o (r.L4-)
LUU
-4
a. 0
0 *U-
CL LC
o-4
u
and the laboratories in such programs as Integrated Communi-
cations Navigation Identification Avionics (ICNIA), Ultra
Reliable Radar (URR), Common Signal Processor (CSP), and the
VIISIC 1750A Data Processor. Module types and sizes are typi-
cal of systems currently being developed. A central element
of the decision analysis is the comparison of line replaceable
units (LRUs), which include subassemblies removable at inter-
mediate maintenance levels, to line replaceable modules (LPJIs)
which are removed at the flight line and repaired at the depot
level. The same VHSIC technology base has been assumed for
either LRUs or LRMs to avoid confusing the technology influence
with partitioning and packaging effects.
Each of the factors to be considered in establishing
a preferred modular architecture is described and decision
-riteria to select alternatives are developed. It is expected
that the baseline established in this study will be used in a
.1 subsequent analysis phase to develop life cycle cost estimates
for the alternatives.
i%
V.,
V.
• 1-3
d#
2. BACKGROUND
Avionics subsystems in today's weapon systems have
become increasingly complex and sophisticated in order to meet
ever increasing requirements. As the role of avionics in crit-
ical areas of flight control, engine control, and mission
performance increases so does the requirement for avionics
reliability. The consequence of electronic failure in flight
and engine control applications may be loss of the aircraft.
The digital flight control processor in the F-15E, for example,
is triplex redundant. The flight control electronics for the
most recent F-16 is quad redundant. Today's fighter aircraft
avionics suites average in the range of 15 to 35 flight hours
between critical failures, according to AFM 66-1 data with an
average time to repair a failure of over two hours. Today
approximately one quarter of the total aircraft maintenance
man-hours per flying hour are spent on avionics problems.
Both the F-15 and the F-16 aircraft rely upon an avionics inter-
mediate shop (AIS) at the intermediate maintenance level. In
future systems, the AIS may be too vulnerable and inflexible
for the predicted support environment.
Air battle planners see the trend for an increasingly
important role for avionics continuing into the mid-1990s, but
in a support environment that is expected to change. Planners
* anticipate the use of a few main support bases from which air-
craft will deploy to operational bases, with subsequent deploy-
ment from the operational bases to dispersed operating locations.
Off-equipment maintendnce requirements must be minimized in
.this mobile, flexible support scenario. Unfortunately, an AIS
-d
is large and expensive, and over four C-141s are required to
deploy the AIS for one F-16 wing and the F-15 AIS is even
larger. Since an AIS is large, expensive, and vulnerable,
reliance on it must be reduced for future aircraft. Planners
predict improvements in mean flying hours between critical
failure in the hundreds of hours and avionics availabilities
of 95 percent or better if future sortie reouirements are to
be met. The mean time to repair must be reduced from two hours
to fractions of an hour if the projected surge rate of up to
five sorties per day is to be achieved.
Modular avionics architectures have been offered as
part of the solution to meet the needs for both high relia-
bility and reduced reliance on flight line support equipment.
Modular avionics architectures have included many different
features designed to reduce life cycle cost, improve mission
reliability, improve maintenance efficiency, or eliminate the
AIS. Each of these factors must be evaluated individually to
estimate their contribution to the goals of the overall avi-
onics concept in comparison to their cost of implementation.
The use of line replaceable modules stems from Navy
initiatives to simplify their support problems with shipboard
electronics. Their Standard Electronic Modules (SEMs) were
designed to be low part count building blocks from which most
electronic functions could be built. When the small building
block failed, it was thrown awav. The Navy SEM concept offered
standardized building blocks which could reduce the spares
0] complement needed to maintain the shipboard electronics. This
concept had two principal shortcomings that did not make it
attractive for airborne use. Because the modules were small,
m packaging density and circuit board efficiencies could not be.achieved, and the weight and volume requirements were excessive.
Since the module circuitry was standardized, technology advances
"32-2
could not be expeditiously implemented and design stagnation
resulted. The Aeronautical Systems Division, Directorate of
Avionics Standardization and Systems Architecture prepared a
module development and implementation plan in the mid 1970s
that set some ground rules for a USAF module program that
stressed the need to allow for technological advance and the
need for efficient functional partitioning. The ground rules
are still appropriate for application today. Advances in semi-
conductor integrated circuit technology that have resulted in
Very High Speed Integrated Circuits (VHSIC) may make it possible
to efficiently partition functional elements at the module
level. Although much has been written about the benefits of
VHSIC LRMs, few cost/benefit analyses have been performed
because of the uncertainty in predicting both costs and relia-
bility of the technology. Now that design efforts are underway
using Phase I VHSIC technology, some cost and reliability pro-
jections are available. Advances in technology may be accom-
modated if standardization is limited to form, fit, function
and interface. The circuitry and techniaues used to implement
the function are not controlled.
Program Management Directive (PMD) 5059(1)63253F/3003,
dated 25 January 1985, directed the development and production
of a Standard Modular Avionics System Architecture (SMASA) for
incorporation in Air Force weapon systems. Key factors in
this architecture included the use of form, fit, function, and
interface standardization of line replaceable modules, on-board
test capability, time stress measurement devices (TSMDs), the
use of the Integrated Maintenance Information System (IMIS),
and management software to allow dynamic reconfiguration of
module tasks in flight. Each of these factors is intended to
improve mission reliability and r luce life cycle cost.
2-3
An evaluation of their interactions and cost is nec-
essary to allow a decision maker to choose the appropriate mix
of alternatives which can achieve the necessary objectives of
reduced maintenance requirements on the flight line, greater
availability, and better mission reliability, while reducing
the life cycle cost of the fielded avionics suite. Hardware
reliability is the most important factor involved in achieving
the objectives established for avionics in the mid-1990s. If
the reliability is sufficiently high, the maintenance needs
and consequently the operation and support costs are reduced.
Increased reliability must be achieved during the early design
phase. A thorough understanding of the mission and mainte-
nance environment is necessary and adequate funding of the
design for the environment must be included in the program
planning. When the best possible hardware reliability has
been achieved, the options open to the decision maker include
the minimization of the maintenance burden through incorpora-
tion of a mix of the factors discussed in this cost/benefit
study formulation.
42-
3. STUDY DESCRIPTION
The structure of an avionics suite in a weapons sys-
tem is a result of many design decisions throughout the
program formulation and development. Selection of the optimum
technical concept for modular avionics may not be possible
because of the uncertainties in factor interactions and the
changing needs of the decision makers. The selection of a
preferred concept based upon the best information available
should be the goal. The results of the cost/benefit studyshoul~d aid the decision maker in selecting the preferred con-
cept for modular avionics.
A systematic approach for identifying and selecting
alternatives must be used in formulating a cost/benefit study.
The modular avionics architecture in this study is in an early
conceptual phase, so many factors which influence the concept
selection and the system life cycle cost are subject to change.
Table 3-1 includes a list of factors discussed in the back-
ground studies and identified in the SMASA PMD of January1985. This list is by no means an exhaustive list of factors
involved in an overall program, but it represents salientfeatures and alternatives that must be considered in the for-
mulation of the cost/benefit study. Table 3-1 organizes the
factors into five categories representing major program life
0 cycle cost elements. Each of the factors is discussed in the
following sections.
3-1
TABLE 3-1
FACTORS INFLUENCING A MODULAR AVIONICS ARCHITECTURE
Program Management
Utilization of a Central Certification Facility
Configuration Management Library
Acquisition and Development
Droduct Performance Agreements
Competition
Design
Functional Partitioning to LRN or LRU
Standardization
Time Stress Measurement Device Incorporation
Inclusion of a Dedicated Maintenance Processor
Testability/Off-Equipment/On-Equipment Test Alternatives
Reconfiguration/Software Development Needs
Operation
Deployment Requirements
Base Locations
Weapon System Quantities
Support
Use of Modular Automatic Test Equipment
Use of Integrated Maintenance Information System
Location of Depot Repair Facility
Selection of Maintenance Strategy
q' Scheduled Vs. Unscheduled Maintenance
4 3-2
3.1 PROGRAM MANAGEMENT1
3.1.1 Certification Facility
A central certification facility is beneficial inmaintaining control of unit configuration and initial design
quality when a number of manufacturers are required to design
to a standard. The facility may be helpful in maintaining an
efficient quality assurance program which leads to high module
reliability. The need for the facility is independent of the
decision to partition into LRMs or LRUs. The decision to uti-
lize a certification facility, Government-manned or contractor
manned, will depend upon the estimated cost in relation to the
improved reliability, the gain of central data control, andthe projected decrease in life cycle cost. The need to centra-
lize control increases as the degree of standardization increases
* and the ability to efficiently and unambiguously test for com-pliance with the standard decreases. The costs for introducing
a certification facility may be estimated separately for in-clusion in a final decision matrix.
3.1.2 Configuration Management Library
Maintaining close control of unit configuration may0 be necessary where standardized units are used in many appli-
cations. A central configuration management facility couldprovide a useful service in maintaining a library of standard
unit ,technical descriptions available for use by system de-signers. The library items maintained could include interface
descriptions, qualification and performance data, and electri-
cal schematics. The data could be maintained in a selectedComputer Aided Design (CAD) format to serve the user needsmost efficiently by allowing remote electronic access to the
information. The library items would not necessarily be re-
2 quired for use on all programs but their use may be encouraged
3-3
bv contract. The configuration management facility could be
maintained by the government, or it could be maintained under
contract to the government organization responsible for modular
avionics. The alternative would be to rely upon existing pro-
fessional societies or industry associations to voluntarily
monitor and disseminate standard technical descriptions. The
cost of these optional strategies can be estimated and included
in the overall cost summary.
3.2 ACQUISITION AND DEVELOPMENT
3.2.1 Product Performance Agreements
Product performance agreements (PPAs) including war-
ranties, guarantees, and other contractual devices are required
by federal law for most major weapon systems. The PPA is used
to provide incentive to the contractor to meet or exceed per-
formance requirements. The PPA should be tailored to the
needs of the specific program. For example, if the risk associ-
ated with meeting performance is high, the PPA can be struc-
tured to offer incentives to meet the performance. Conversely,
if the hardware design is an off-the-shelf standard unit, a
complex reliability improvement warranty may not be cost effec-
tive. The cost of the PPA must be considered in the life cycle
cost estimate.
The Product Performance Agreement Center's (PPAC's)0 automated Decision Suppnort System (DSS) was used to determineJ.
potential warranty approaches tor this program. Warranty se-
lections were made based on prcyram -biecives and equipment
characteristics as they are known at this time. The list of
potential warranties inclided:
0 -
0 Reliability Improvement Warranty (RIW)
* Mean Time Between Failure VerificationTest (IHTBF-VT)
* Logistics Support Cost Guarantee (LSCG)
* Incentive Award
* Reliability Guarantee
* Availability Guarantee
* Component Reliability Guarantee
* Interim Contractor Support
* Chronic LRU Guarantee
* Repair/Exchange Agreement
* Correction of Deficiencies
" Commercial Service Life.
3.2.2 Competition
A potential factor in reducing acquisition cost and
in maintaining quality in subsequent avionics procurements is
competition. Competing companies will hold their proposed
costs down and will maintain a cost effective production facil-
ity when follow-on buys are expected to be competed. Product
quality will receive greater emphasis if the result of poor
quality is loss of follow-on business. To have competition,% multiple sources of supply must be available and willing to
0 bid on the initial contract and on subsequent contracts.
Standardization of requirements can increase the order size
and, consequently, attract multiple potential sources. Formn,fit, and function standardization allows alternative design
solutions to be proposed which can foster competition while
reducing the possibility of technological stagnation. Coin-
~~petition may be fostered by awarding multiple contracts for
S 3-5
_. a -
hardware development and for subsequent production. Leader-
follower contracting Cp~rOpr( ihcs can b( used to retain competi-
tion in subsequent huvs of the -,vionics assemblies.
3.3 DESIGN
3.3.1 Partitioning
Partitioning in a general context, is the allocation
of circuit functions to separately identifiable assemblies or
software modules. In general, the life cycle costs are not
expected to be sensitive to moderate changes in module or
printed wiring assembly size, unless those changes impact the
way the units are supported. For the purposes of the study we
are most interested in the partitioning of the functions to
liae replaceable units containing separate shop replaceable
units or in the partitioning of functions to line replaceable
modules that are not expected to undergo further disassembly
at intermediate level. The differences between LRUs and LRMs
should be apparent in the life cycle cost comparisons.
Included in the design category are a number of factors
that represent critical decision points in selecting a preferred
avionics concept. The primary decision centers on selection
of an LRM or an LRU for the basic hardware building block of
the avionics architecture. If the acquisition cost of the
line replaceable element can be kept low and the reliability
4 can be made to be high, significant reductions in operation
and support costs are possible as suggested by Dougherty in
Reference 1. His proposed criteria for the elimination of the
intermediate maintenance level is shown in Fig. 3.3-1. If the
acquisition cost per failure free operating hour lies to the
- 3-6
I.
A-32415
1,000000
c 10,00 -
. INU
S10,000 RADAR/FIRE CONTROL
8 0
0
U) 0
A z 1,000o
-~. "0
0 ./ F-16 AVIONICS SRU'S FORCU " 3-LEVEL LRU'S
100
10 1 I I
10 100 1,000 10,000
,. MEAN TIME BETWEEN REMOVALS (HOURS)
. Figure 3.3-1 Evaluation Criteria for DeterminingA> Maintenance Level
I
right of the trend line, use of a two-level maintenance strategy
p.2- is cost effective. To the left of the trend line, a three-level
maintenance strategy should be employed.
* .3-7
'I-&
Reductions in component count by partitioning to
smaller line replaceable elements are expected to increase
availability and reduce costs. Further benefits are assumed
as the design is partitioned to smaller functional elements
which lend themselves to standardization. The costs for both
LRU and LRM alternatives should be fully developed in the
cost/benefit study.
3.3.2 Standardization
The degree of standardization is a significant factor
in life cycle cost determination. The interoperability
benefits of interface standardization using the MIL-STD-1553
multiplex data bus are recognized by industry and have been
documented in the technical literature. Form, fit, and func-
tion standardization at the LRU level has been used with
success in avionics applications such as the standard inertial
navigation system and the standard central air data computer.
Although standardization down to the circuit design level
would offer some reductions in support costs, the price paid
in technology stagnation and limiting competition may be large.
The optional use of elements common within a weapon system may
reduce the operation and support cost through reduction in the
spares types needed, without the interoperability concerns
which accompany the use of standardized hardware. It may be
possible to realize some design cost savings by using common
modular elements, but the savings may be offset by the burden
of constraining the designer to conform to the limitations of
available common elements. The cost of common modules will be
developed for the LMI alternative only, since an analogous
common LRU does not appear to be practical.
3-8
3.3.3 Testability
Testability is an important element in the definition
of a modular avionics concept. The percentage of off-equipment
to on-equipment testing should lean heavily toward on-board
test (OBT) in order to reduce the requirement for externalsupport equipment. OBT should reduce the overall life cycle
cost while improving operational effectiveness. OBT can pro-
F"'. vide an indication of performance to the aircrew as in the
case of an initiated self-test during preflight. OBT can also
provide fault prediction, detection and isolation for mainte-
* nance. The potential for prediction in avionics has not yet
been fully explored. Detection and isolation must be taken to
the line replaceable element in order for OBT to be effective.
The importance of OBT detection and isolation functions in-
crease as fault tolerance through dynamic reconfiguration of
hardware task assignments is introduced. Intermittent or non-functioning OBT will adversely affect reconfiguration capability
and reduce mission reliability. OBT can be expected to utilize25 percent or more of module surface area even with the on-chip
test capability of Vi-SIC fully utilized. The selected testa-bility strategy will have a significant effect on life cycle
cost, both in terms of initial development and operations andsupport. The strategy may include a dedicated maintenanceprocessor to track information on intermittent failures, the
- redundancy levels used, history of reconfiguration, and envir-
onmental exposure history.
3.3.4 Time Stress Measurement Device (TSMD)
The TSMD is a stress and time sensor integral to the
electronic equipment. The purpose of the TSM is to sense the
environment for recording and later use. Usage can vary from
test instrumentation for initial qualification to continuous
* 3-9
real time monitoring of the environment. Initial devcopment
work by Battelle Columbus Laboratories and Minneapolis Honeywell
has demonstrated the feasibility of the concept. Work still
remains to reduce the size of the TSMD and to develop an overall
OBT strategy which includes the TSMD. It is now possible to
consider incorporating a TSND in each line replaceable unit as
the required area is reduced using VHSIC technology. Storage
of data is still expected to pose a problem. Since the memory
-available to dedicate to this storage task is expected to be
limited, only selected data should be recorded such as parameter
-. excursions above a threshold. Temperature, temperature rate,
* vibration, and power are expected to influence the electronics
and consequently should receive recording priority. For vibra-
tion, the spectrum must be divided into segments. Excursions
above preset thresholds in each of the segments would be re-
corded. The TSMD system includes not only the sensor embedded
within the equipment, but also the data recording and data
reduction equipment required.
3.3.5 Reconfiguration/Software Development Needs
The prototype architecture for the modular avionics
cost benefit study is a PAVE PILLAR distributed, reconfigurable
multiprocessor architecture. An architecture of this general
form could be implemented using LRUs or LR1s. As noted previ-
ouslv, it is necessary to break an overall strategy into com-ponent parts and to make simplifying assumptions in order to
evaluate alternative concepts for a multifaceted program such
as this. One such component part is the real-time operating
system software needed to tie the application dependent soft-
ware modules together, and to further control the reconfigura-
tion of the system upon failure of a host processor or inter-
face. The assumption is made that an alternative approach
would allow the application dependent software modules to
0 •3-10
4. 4.0
operate autonomously in dedicated processors as in a conven-tional fighter weapon system of the mid-1980s. The differencein cost will be the development and maintenance cost of themanagement software. The potential benefit will be derivedfrom the increase in mission reliability brought about through
dynamic reconfigurability.
3.4 OPERATION
The operational factors influencing the selection ofa modular avionics architecture include weapon system quantities,base locations, and deployment requirements.
It is assumed that many current weapon system charac-
teristics are adequate to evaluate systems to be deployed inthe mid 1990s. The operational factors are discussed in detail
in the section on scenario development, so no additional dis-cussion of this factor will be included here.
3.5 SUPPORT
e' 3.5.1 Modular Automatic Test Equipment (MATE)
If external support equipment is required for offequipment maintenance, the option exists to use MATE. Thealternative is to develop that equipment tailored to the needs
of the modular avionics concept. Since use of MATE is not a
central issue in the study being formulated, it will be evalu-
ated separately so the costs may be included in an overalllife cycle cost analysis. It is anticipated that with OBT,
0 off-equipment maintenance needs will be minimized.
* 3-11
3.5.2 Integrated Maintenance Information System (IMIS)
Research has been sponsored by the Logistics and
Human Factors Division of the Air Force Human Resources Labo-
ratory to develop a means to integrate and deliver automated
flightline maintenance information from various sources. THIS
makes use of a portable computer to display graphical technical
data, to interrogate on-board systems, and to analyze
in-flight failure information. The IMIS will provide
intelligent diagnostic advice for maintenance personnel andTHIS should reduce the maintenance personnel needed to perform
the wide range of maintenance required in the tactical
0environment. The need for IMIS is dependent upon the
testability concept adopted. If the maintenance is simplified
* through use of on-board test, flight line maintenance
information may not be required, but the fault information
downloaded from the aircraft systems may be very useful in* -V speeding turnaround of aircraft on the flight line.
3.5.3 Depot Repair Facility
A key consideration for selecting a depot repair facil-
itv is the use of a contract facility versus a Government facil-
ity. With the increase in reliability of hardware and sophisti-
* cation of testing anticipated with VHSIC modular avionics, it
is possible that contracting for depot repair may be a viable
option. The repair facility could be made part of the certifi-
cation facility. A second option is to contract for repairs
by the vendor using a warranty agreement. Cost estimates for~s. the contracted depot repair should be developed for evaluation
of alternatives. This optional factor is independent of the
central decisions associated with selection of LRMs versusLRUs so it may be evaluated separately and added to the results
at the conclusions of the study.
3-12
3.5.4 Maintenance Strategy
Some factors regarding maintenance strategy have been
previously discussed, but one additional factor associated with
the study is the decision to use scheduled versus unscheduled
maintenance. Unscheduled maintenance or maintenance on-demand
has been the normal maintenance mode for avionics. As relia-
bility of hardware elements is increased and the understanding
of the mechanisms of failure improves, scheduled maintenance
becomes possible. Maintenance on-schedule can reduce the AIS
requirements and improve mission reliability. If the causes
of hardware failure have been isolated, expected failure free
time can be predicted, allowing for depot-only maintenance of
line replaceable elements on a predetermined schedule. The
cost and potential benefits for this factor may be developed
separately for optional inclusion in a final strategy.
3.6 STUDY ORGANIZATION
The many factors influencing the selection of a
preferred modular concept have been individually discussed.
Each of the factors is intended to contribute to improved
mission reliability and reduced life cycle cost. There is
interaction among some of the factors while others are inde-
pendent. Figure 3.6-1 shows the relationship among the
primary or core factors through use of a decision tree.
4 In order to provide a logical basis for selection of
a preferred concept, it is necessary to first establish the
criteria upon which the selection is to be made. This estab-
lishment of decision criteria is discussed in the next chapter.
3-13
A -32406
DEDICATED
APPLICATION-E ;UNIQUE SOFTWARE
MANAGED
LRU RECONFIGURATION
BASED
DEDICATED
STANDARD SFWR
MANAGED
COST EFFECTIVE D ED ICATEDAVIONICS ARCHITECTUREI
APPLICATIONUNIQUE SOFTWARE
MAAERECONFIGURATION
DED ICATED
LRM COMMONL RMWITHIN L
BASED APPLICATION OFWARE
RECONFIGURATION
I DEDICATED
p STANDARDACROSS SOFTWARE
APPLICATION MANAGEDRECONFIGURATION
Figure 3.6-1 Modular Avionics Decision Tree
* 3-14
- ~.4. DECISION CRITERIA
The decision criteria by which an alternative is
selected for inclusion in the preferred concept are developed
and discussed in this chapter. The discussion is organized to
follow the five major cost categories introduced in the previ-
ous chapter.
* 4. 1 PROGRAM MANAGEMENT
4.1.1 Certification Facility
Figure 4.1-1 shows the decision tree for this factor.
A-3241 2
COMPLIANCEDEMONSTRATED
-~ ByMANUFACTURER
LINE REPLACEABLEELEMENT GOENMN
*SPECIFICATION FACILIMTCOMPLIANCE FCLT
CENTRALCERTIFICATION
FACILITY
CONTRACTEDCERTIFICATIONINDEPENDENT
ACTIVITY
*4Figure 4.1-1 Certification Facility
4-1
01%
Criteria - Cost, Government manpower, facilities/equip-
ment, risk of non-compliance, risk of degraded quality assurance.
Discussion - The expected contribution to life cycle
cost for a certification facility will be in manpower, facili-
ties, equipment, and test program development. The Naval Weapons
Support Center (NWSC) facility at Crane, Indiana is an example
of what may be required to support a certification program.
This facility occupies about 20,000 sq. ft. and has a staff of
about 100 people. They presently test a combination of about
3000 SEM format A and B modules a year. Facility size and
manpower will be dependent on the tasks to be performed such
as the following accomplished by NWSC:
Specification Development
Design Verification
Initial Qualification
Periodic Check
Test Procedure Correlation
Manufacturer Audits
Change Control
*Qualified Products List Maintenance
Failure Analysis.
Also to be considered is the type and extent of testing
to be performed. Some of the factors to consider are:
Functional Tests - Digital and analog withautomatic test equipment and bench equipment
Environmental Tests - Temperature, humidity,vibration, thermal shock
4-2
k"On
Failure Analysis - Scanning electron microscope,low power optics microscopes, disassemblytools
Test Development - Test jigs and software.
Some of the benefits to be gained are reduced risk of
" introducing a non-compliant element into inventory and a cen-
tralized data pool for element specification. A more important
benefit to be gained with an extensive Quality Assurance Program
would be higher module reliability and consequently a lower
life cycle cost.
I
The investment in equipment for a facility similar to
that at NWSC is estimated to be $10M, but the pay back is in
improved production techniques, standardized test procedures,
reduced development and test costs, which result from central-
ized development control. The alternative permitting the manu-
facturer to demonstrate compliance, will not introduce cost,
but the potential for allowing non-compliant hardware into the
inventory increases. The need for consideration of this factor
increases when standard line replaceable elements are used.
4.1.2 Configuration Management Library
The decision tree for this factor is shown in Fig.
4.1-2.
Criteria - Cost, facilities, manpower, proliferation
of configurations.J
Discussion - If there is no ready source of standard
unit data, compliance with standards may be lost and designers
would have no way of determining if a standard unit fitting
the design requirement is available. The various options have
4 4-3
Li~a..~d
A 324 11
VOLUNTARYMONITORING
OFCONFIGURATION
CONFIGURATION GOVERNMENTMANAGEMENT MAINTAINED
CENTRALCONFIGURATION
CONTROL
CONTRACTEDINDEPENDENT
CONFIGURATIONMANAGER
Figure 4.1-2 Configuration Management Library
differing implemention costs and varying probabilities of suc-
cess. The costs for each alternative are to be developed.N This factor is applicable only to concepts incorporating stand-
ard line replaceable elements.
If a data library is maintained, the cost of develop-
ing it and maintaining it must be evaluated with regard to the
potential benefit to be derived. The benefit is not directly
quantifiable in terms of savings to the Government. It is a
service to the design community to facilitate the use of thestandard units and to make their use the most cost effective
* procedure for the hardware designer. The Government can expectto realize cost savings through the widespread use of the stand-
ard design information. Questions to be resolved before estimat-
ing the cost of implementation include the formatting and release
*of the data. A Computer Aided Design (CAD) data base would
offer significant advantages including easy access and efficient
updating. The library facility could be small. The librarystaff would be required to update and maintain the file data
on the computer. The cost of establishing and maintaining the
faciitvby the Government, under contract to the Government,
or by an independent organization should be developed.
44-4
4.2 DEVELOPMENT AND ACQUISITION
4.2.1 Product Performance Agreements
The decision tree is shown in Fig. 4.2-1.
A 32410
LIMITEDMATERIALS ANDWORKMANSHIPWARRANTY
RELIABILITY%PPA IMPROVEMENTWARRANTY
PERFORMANCEINCENTIVES
- OTHER PPA
Figure 4.2-1 Product Performance Agreements
Criteria - Cost, performance, reliability.
Di-cussion - Due to the development status of this
program there is insufficient data to support an in-depth quan-
.7 titative analysis, although the various warranty approaches
were reviewed for their applicability. For instance, any of
the warranties which stimulate reliability growth, such as the
-. RIW, appear too risky for this program. The basis for this
,'.. reasoning is that new advanced technology which has no field
or prior operational experience is likely to render the RIW a
high risk option due to the high level of risk in estimating
present reliability levels, reliability growth, and failure
trends. There are also risks in the lack of knowledge about
the operational environment, installation, and integration.
.'. The Government, and also the contractor in this case, should
be aware of the risks and conduct further analysis to determine
*o 4-5
" 1.
if an PIW is applicable. If an RIW is applicable, care should
be taken to minimize the risks.
On the other hand, a warranty such as an MTBF-VT has
* reasonable applicability to the connon module program. Gene-
rally, under the terms of an t!TBF-VT the contractor guarantees
a certain initial level of field MTBF. The field performance
is determined by a verification test conducted by the user
according to a testing scheme prescribed by the warranty.
Verification test periods might begin six months after the nth
unit is delivered and conducted at six month intervals during
the warranty period. The guaranteed MTBF is usually increased
, during each test period so that the ultimate reliability goal
is reached at or shortly before warranty expiration. When the
contractor fails to meet a measurement goal, the Government
typically receives consignment spares (at no additional cost)
to improve system availability.
With the limited data available it appears that the
MTBF-VT or a similar approach will meet the needs of the common
module program. In addition to the basic warranty, a Correction
of Deficiencies clause and a warranty on Materials and Workman-
ship will be required to fulfill the requirements of the law
as stipulated in the 10 U.S. C'de Section 2403 (10 USC 2403).
It is possible to estimate warranty cost on the basis
4.. of historical costs of avionics programs using advancing tech-
nology. At this point, the common module program appears to
O be a moderate to high risk program from a warranty perspective.
Accordingly we estimate warranty cost at seven percent of re-
curring unit cost per year.
J.
; -,
.. .,., - , ...-.. ..0.-,., ... . .. . ,. .,., . ., ,., ., , , ., .> , ,
4.2.2 Competition
The decision tree is shown in Fig. 4.2-2.
MULTIPLE CONTRACT AWARD
COMPETITION ILEADER-FOLLOWER CONTRACTSTRATEGY
C 'UALIFIED PRODUCTS LIST
, Figure 4.2-2 Competition
C-riteria - Acquisition cost, quality, and production
.apacitv.
Discussion - The costs of competition vary with the
strategy employed. Logistics support costs will increase as
the number of suppliers providing form, fit, and function com-
patible hardware that is not physically identical is increased.
Tooling and test equipment such as interface test adapters and
'est program sets must be provided for each supplier's hardware.
Government management costs increase with the number of separate
contracts that must be managed. In the leader-follower contract
case, the leader's hardware design is provided to the follower.
The cost of providing the design information to the follower
* and follower startup costs must be assumed by the Government.
The benefits to be achieved through competition include reduced
acquisition costs and greater surge capacity in industry if
production rate increases became necessary. Each individual
* supplier is more aware of his production efficiency and gener-
"- allv makes more cost effective decisions in the face of compe-
tition. Experience has shown that cnmpetitors are willing to
make capital investments and introduce innovation into the
production line to improve their competitive position. Quality
and reliability of the product are more important to a manufac-
turer who may lose subsequent business if his product is not
0 1W %
as durable as that of his competitor. There is no single stra-
tegy for competition, just as there is no single product perform-
ance agreement that is ideal for all procurements. The strategy
must be evaluated with respect to the overall program objectives.
- 4.3 DESIGN
4.3.1 Partitioning
Figure 4.3-1 shows the decision tree for this factor.
The criteria, shown in Table 4.3-1, used in evaluating the
alternatives are discussed in the following paragraphs.
A 32408
LRM
OPTIMUMLINE
REPLACEABLE-~ ELEMENT
LR UWITHSRUs
-, Figure 4.3-1 Partitioning
Discussion - An important constraint associated with
partitioning is the need to detect and isolate faults to the
module or unit during testing. If the ambiguity group includes
0 more than one module, then more than ore module must be removed,
replaced, and sent through the repair pipeline. This constraint
would tend to keep the module size sufficiently large to assure
-" that the ambiguity group is no greater than one. As Fig. 3.3-1
*indicates, the cost/failure-free operating hour must be kept
b" .8S"
o*. .
TABLE 4.3-1
PARTITIONING DESIGN CRITERIA
Program Management
Manpower Costs
Acquisition and Development
Development Cost Acquisition Cost. Industry Support Competition
Schedule/Risk ProducibilityInterservice Applicability
Design
Space/Weight/PowerHardware ReliabilityTechnology IndependenceTestability
Fault Isolation and DetectionFault Tolerance
PackagingThermal Management
Corrosion ControlChemical Biological Warfare ProtectionElectrical Overstress/Electrostatic Discharge ProtectionMounting Provisions
Operation
Mission ReliabilityEase in Adapting to New MissionsAvailabilityDeployment CostsMission Performance
Support
, Spares Maintenance
Storage ManpowerHandling FacilitiesItem Management Support EquipmentData Management Training
Technical DataMaintenance
0 4-9
-4 -- - - - - - - - - - - - - - -
* low to hold down the spares cost. If the unit is high in cost
and low in reliability, the investment in spares to keep the
pipeline flowing raises the life cycle cost. If the reliability
of the module is sufficiently high, fewer spares are needed in
the pipeline and the life cycle cost is reduced. These con-
straints tend to reduce the size of the line r--placeable element
in order to reduce complexity which in turn would reduce element
cost and increase element reliability. Partitioned element
size is also constrained by the information flow requirements
across the element boundaries. Speed and noise constraints
will tend to keep the elements sufficiently large to contain
the high speed, low noise interfaces on a single element.
Both VESIC and monolithic microwave integrated circuitry are
allowing the hardware size to shrink so that functions can be
contained on practically sized elements.
Program Management - There is no apparent difference
in cost between management of an LRU program or an LRN program
if the standardization factor is not considered.
Development and Acquisition - The costs for develop-
ing and acquiring LRUs or LR~s should be similar. Packaging
constraints associated with protecting the smaller LRNs may
introduce added costs however. Assuming that the same level
4 of technology is used for both the LRU and the LRd4, there is
no anticipated difference in risk. The smaller the functional
element, the greater the potential for multiple uses within
and between services. Given the same technology, either
approach should be producible. Competition can help to hold
acquisition costs down when the procuring agency continues to
allow prospective bidders to propose solutions to requirements,
rather than requiring exact design duplications.
4-10
% e%%
Design - More space, weight and power may be required
for an LRM implementation owing to the increase in individual
module volume and weight to protect it in the handling environ-
ment. Hardware reliability may be reduced by the additional
circuitry needed to implement OBT to the module level. For
the LRU alternative, OBT must isolate only to the unit level
unambiguously. Fault detection and fault isolation for both
LRUs and LRMs are expected to be aided by the on-chip fault
detection capabilities offered with VHSIC. Packaging con-
straints for the LRM are expected to add volume and weight to
the overall functional element. Thermal management and cor-
rosion control techniques are implementable for either LRUs orIL R-Ms. Chemical/biological warfare protective measures are
similar to corrosion control techniques in many respects when
module provisions are compared. A conformal coating like
PARALENE has been found to be effective in many cases but it
is hard to remove for repair. In both the LRU and LRM con-
figuration, the exposed interface connectors are vulnerable
when the element is removed from the weapon system. Electrical
overstress and electrostatic discharge protection may require
bypassing/filtering of pins on the exposed connectors which
would penalize the LRM more since there are more exposed
element connectors to be handled than would be the case with
an equivalent subsystem LRU.I
Operations - From an operations point of view, the
alternative with the most potential for eliminating the avi-
onics intermediate shop must be given the advantage. Without
0considering the support costs, the reduced dependence on the
AIS reduces vulnerability. It facilitates airfield deployment
by eliminating the airlift requirement for the AIS. There is
no apparent performance difference between LRUs and LRfs. The
potential for mission reliability improvemenc increases as the
4-11
number of common elements subject to interchangeability inflight increases. Both LRUs and LRMs can be configured to
allow for redundancy but the potential for success is greater
using LRMs, since the smaller functional elements are easier
to replicate and use as common elements in a weapon system.
Support - Support costs are heavily influenced by the
acquisition cost and reliability of the spare line replaceable
elements. Since the LRMI is smaller, the cost should be lower
:- and the reliability higher, giving it an advantage in the sup-
port area.
* K' 4.3.2 Standardization
Figure 4.3-2 shows the decision trees for this factor.
It should be noted that standardization can be used with LRMs
or LRUs, therefore, costs must be developed for both LRI-Is and
LRUs. The discussion includes the use of common modules within
a weapon system.
UNIQUE A 32413
,LRU un-LR
STANDARD
UNIQUEILRM COMMON
STANDARD
Figure 4.3-2 Standardization Decision Trees
4-12
orq
Table 4. 3-2 shows the decision criteria applied to
this factor. A discussion of each of the five cost elements
is included in the following paragraphs.
Program Management - Standardization brings with it a
need for some centralization of control. Optional development
of a certification facility and a configuration management
library have been discussed separately, but regardless of these
decisions, management costs can be expected to increase for
standardization by the government. Much of the increase will
be due to the added manpower to prepare and monitor the speci-
* fications and standards. Management of common elements within
a weapon system would be the responsibility of the system
integrator and would not affect government resources.
Development and Acquisition - Development costs can
be expected to increase to cover additional compatibility
testing required for standard hardware. Industry support for
standardization is often withheld, or given reluctantly, due
to the perception by industry that standardization by the
government means another tier of requirements documents and a
potential reduction in the opportunities to compete. The
strategy must be developed carefully and then sold to the
avionics industry in order to gain their support. The devel-
opment of standardized hardware can be expected to take longer
since there are more design constraints to factor into the
process. Schedules may be at risk in the development phase.
4 In follow-on acquisition of standardized hardware, the schedule
* risk should be less because the requirements have been wellestablished and multiple sources for elements meeting those
requirements are available. The potential exists for inter-
service applications where standardized requirements have been
developed with the cooperation of the services. The smaller
the functional partition of the elements, the more likely that
interservice applications can be found.
4 4-13
K i
,
TABLE 4.3-2
STANDARDIZATION DECISION CRITERIA
I Program Management
" - Manpower Costs
-Acquisition and Development
Development Cost Acquisition CostIndustry Support CompetitionSchedule/Risk ProducibilityInterservice Applicability
Design
* Space/Weight/PowerHardware ReliabilityTechnology IndependenceTestability
Fault Isolation and DetectionFault Tolerance
PackagingThermal ManagementCorrosion ControlChemical Biological Warfare ProtectionElectrical Overstress/Electrostatic Discharge ProtectionMounting Provisions
Operation
Mission ReliabilityEase in Adapting to New MissionsAvailabilityDeployment CostsMission Performance
Support
Spares MaintenanceStorage ManpowerHandling FacilitiesItem Management Support EquipmentData Management Training
Technical DataMaintenance
4-14
- ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ X-m a- -r , -- r rS wrr n n- -
Producibility is a result of careful consideration of prod-uction needs during the design phase. Standardization of
functional requirements may result in more mature designrequirements meaning fewer last minute design changes where
producibility may not be adequately considered.
Design - Standardization of hardware form, fit, and
function implies that some design aspects must be compromised.
Technology independence is maintained if standardization isrestricted to form, fit, function and interface. Testability
-4 and packaging design become more complicated as requirements are
introduced for standardization, but the effects are expected
to be minor. The design of common elements within a weapon
system offers the system integrator some of the advantages ofstandard hardware without the restrictions brought about bycompatibility requirements for other systems. Design costs
for the weapon system should be reduced with common elements.
Operations - The use of standard or common elements
has little direct effect on hardware performance, consequently
their use has little direct effect on mission performance.
Support -Common and standard flight hardware allows
standard support equipment to be used. Savings in training
0 brought about by the reduction in modules and reduced inter-N mediate shop needs can be expected. Since the total number of
elements is reduced through standardization, spares quantities4' will be reduced which will reduce all aspects of support cost
0 including storage, handling and management.
4.3.3 Testability
The decision regarding testability is one of implemen-
tation. The decision tree in Fig. 4.3-3 has been simplified
to show two alternatives, OBT versus off-equipment test. In
4-15
9 O
fact, a combination of the two alternatives is a more likely
outcome. OBT may be implemented with a software intensive
solution or a hardware intensive solution. The hardware inten-
sive solution will have the greater impact on physical charact-
eristics like size, weight, and power. The software intensive
solution may impact processor timing and sizing unless a dedi-
cated processor is added.
A-3241 4
ON-BOARD TEST
TESTABLILITYIMPLEMENTATION
LOFF-EQUIPMENT TEST
Figure 4.3-3 Testability
Table 4.3-3 shows the decision criteria for the
testability factor. Each of the five categories are discussed
in the following paragraphs.
Program Management - No impact to program management
cost is anticipated since the testability features are in-
cluded in the overall hardware/software design effort for OBT.
Off-equipment testing requirements may generate additional
support equipment needs that must be tracked.
Development and Acquisition - As the testabilityburden shifts to OBT, prime equipment development and acquisi-
tion costs can be expected to increase. Both hardware and
software will be effected.
-' 4-16
Im_ Ai '
TABLE 4.3-3
TESTABILITY DECISION CRITERIA
Program Management
Manpower Costs
Acquisition and Development
Development CostSchedule/RiskAcquisition CostCompetition
'" Design
Space/Weight/PowerHardware ReliabilityTestability Characteristics
Performance AssessmentFault PredictionFault DetectionFault Isolation
Operation
Mission Reliability
AvailabilityMission Performance
Support
Maintenance
9ManpowerI* Facilities
Support EquipmentTrainingTechnical DataMaintenance
Design - Space, weight, and power requirements will
increase as a hardware solution to OBT is adopted. Estimates
range from five to 35 percent of the available board space
'. dedicated to OBT. The major functions to be performed by the
. adopted testability concept include:
io 4-17
1. Performance Assessment
Performance assessment of the avionics for the aircrew
is an important task of OBT. The thoroughness of the test,
the reliability of the test equipment, and the time it takes
to complete the test are all criteria to be used in measuring
performance of the concept. Since this performance assessment
is expected to be performed in-flight, it must be performed by
OBT.
2. Fault Prediction
Fault prediction is a task not normally performed for
the avionics suite, but one that may be possible as advanced
techniques are developed. Incipient engine fatigue failure is
predicted based upon environmental exposure. It may be pos-
sible to predict incipient fatigue failures of avionics, based
upon knowledge of accumulated exposure to environments. The
effectiveness of a fault prediction capability may be determined
by tracking the number of faults found between predicted fail-
ure-free intervals.
3. Fault Detection
0 It is important to be able to accomplish fault detec-
tion on- -ard the weapon system when it is used to initiate
reconfiguration. Criteria to be applied in evaluating per-
formance of the concept include the percentage of faults
detected, the fraction of false alarms, and the mean time
between failure. The can-not-duplicate rate, the mean fault
detection time, the mean run time, and the mean time to repair
K are also important criteria.
4-18
.............:
4. Fault Isolation
Fault isolation is the fourth task to be performed bythe testability concept. Criteria to be applied to the isola-
tion task include test thoroughness, fault isolation resolution,
fraction of false removals, fraction of faults isolated, re-
- ~test OK rate, and mean time to repair. Fault isolation mustbe capable of locating the fault to within a line replaceable
element to reduce the ambiguity group that might otherwise
have to be removed on the flight line.
* Operation - Testability impacts operations through
its effect on mission reliability and availability.
Availability is influenced by the mean time to repair which is
-~ driven by the testability concept.
Support - Maintenance manpower and facilities require-
ments increase as the testability conce pt uses off-equipment
testing. Support equipment needs, including training, technical
data, and maintenance are driven~ heavily by off-Equipment test
needs.
4.3.4 Time Stress Measurement Device (TSMD)
The objective to be met by the TSMD system must be
established in order that it can be evaluated with respect to
meeting the objective. Rome Air Development Center (RADC) and
the Aeronautical Systems Division (ASD) have been cooperating
in a program to define how the TSMD system might be used and
to establish what data should be recorded. Some potential
uses for the system include:
%.
% % A
.-*
1. Product Performance Agreement (PPA) Support
This use ma... nit ustifv the capital investment in
data reduction facilities and in the impact on the equipment,
since PPA's are typically applied for the first five years in
the life of a new hardware development program. Performance
or Reliability Incentive Agreements are useful for early design
contracts, hut such agreements would not impact reliability
for follow-on reprocurements of existing designs.
2. A Source of Environmental Data For Use in Improving
0 Future Designs
The reduction of data collected for this use could be
done at a central location and it could be accumulated in batch
quantities since time response is not critical. Information
related to a specific aircraft would not be required. Statist-
ically significant information on the fleet of aircraft could
be gathered from selected instrumented aircraft.
3. An Environmental Data Source Used to Establish HardwareRetest or Replacement Intervals
0 This information is time critical and it should be
correlated to a specific aircraft. For these reasons, the
data reduction and interpretation should be done at each base.
Scheduled maintenance based upon environmental exposure will
* be possible when fatigue failure mechanisms in electronic
assemblies are better understood. The failure prediction cap-
ability could improve uission reliability and availability but
the manpower workload fhr data reduction and interpretation at
base level will increase.
.°., * .
V,.
The planned use of. dtrti 'i"i, dtermine the estimated
costs for implementin,4 7S>:i lvste. The TSMD may be con-
tained in a single chipu - asamit)e of nodules but the memory
storage requirement, the mI etood o: rti ieva 1 of the data, and
the reduction and dis~erina:+>m o: the results must be con-
sidered in the cost of it: -en ation. Availabilitv of envir-
onmental data in isef willi not improve system performance.
The program must include ur visions for using the data to in-
crease the hardware durabilitv through design improvements.
The cost elements for a TSMD data reduction facility
include:
(I) Staff - The base level manpower require-ment is dependent upon the number ofweapon systems being monitored and theform of the information being collected.
(2) Data Reduction Hardware - A playback- device for the weapon system data and a
computer based interpretation systemwill be necessary.
(3) Central Ia!nagement Organization - Anorganization will be required to inter-pret and take action based upon theenvironmental data gathered.
0- The criteria to be used in evaluating these alterna-
tives include the implementation cost- and the expected benefit
. associated with knowledge of the environment. Knowing the
environment after the design has been fielded will allow the
. effect of mission changes, for example, to be determined. It
can be used as a tool to predict hardware reliability or
failure-free periods hetweer L cieduled maintenance. Tf sched-
uled maintenance is not an~tic-_ipated to be an option, the
knowledge of the environment hust he used to idertify weak
designs for product improve::ents.
0
4.3.5 Software Development for a Dvrami'allv ReconfigurableArchitecture
Reconfiguration is a means of improving mission re-
liability. However, the hardware reliability is not improved
and if the maintenance philosophy is such that hardware fail-
ures are to be corrected when they are identified, the mainte-
nance burden may not be reduced and the overall operation and
support costs may remain high. The decision tree for software
development for reconfiguration is shown in Fig. 4.3-5.
A-32417
DYNAMIC*RECONFIGURATION
IMPLEMENTATIONFOR MISSIONRELIABILITY
DEDICATEDHARDWARE
Figure 4.3-5 Reconfiguration Software
The decision criteria used for evaluation of the
alternatives must focus primarily on the life cycle cost of
6 each option. Development costs of the software associated
with the dynamic reconfiguration alternative are expected to
be sizeable, but operation and support costs should be low.
, t Men dedicated hardware is used to achieve required reliabil-
* ,itv levels through redundancy, the redundant hardware must be
supported over the life cycle, . Reconfi7urqt-lon assumes that
the processors are sufficentv ommon -3 allow operational
flight programs to be run interchanpeahlv.
'.
* ,'-2
4.4 SUPPORT
4.4.1 Modular Automatic Test Equipment: (MATE)
The use of MIATE for support equipment needs should be
evaluated with respect to alternative means of achieving the
same results. The decision criterion is the influence on life
cycle cost associated with using MATE. As the testability
concept shifts toward OBT, less intermediate level test equip-
ment is required. As the need for external test equipment is
reduced, the life cycle cost impact of using MTE is corre-
spondingly reduced.
4.4-2 Integrated Maintenance Information System
The decision to implement IMIS must be made on the
basis of life cycle cost impacts to the weapon system in rela-
tion to projected improvements in mean-time-to-repair in using
the automated technical data. The impact on maintenance effic-
iency becomes less important as hardware reliability increases
N, and OBT successfully isolates to the faulty element. The IMIS
concept has yet to be proven in the field so there is somerisk associated with its icroain
*4.4.3 Depot Repair Facility
The decision tree in Fig. 4.4-1 shows the alternatives
-'. to be considered.
The criteria used in selecting the preferred alterna-
tive includes the following:
1. Life cycle cost of staffing and maintain-ing the depot
- 4-23
2. Contract cost for contractor maintenancefacility
3. Cost of returning parts to the vendorfor repair
4. Expected failure rate of the avionicshardware to be repaired
5. Turnaround time on repairs
6. Surge capability in time of conflict.
A 32407
GOVERNMENTFACILITY
DEPOT
LEVEL _REPAIR
REPAIR UNDER* .~.CONTRACT TO
GOVERNMENT
Figure 4.4-1 Depot Repair Facility
The avionics hardware reliability plays a significant
A part on this decision. If the hardware reliability is suffi-
ciently high. the largely fixed cost of the depot repair facility
could be higher than the equivalent repair capability performed
under contract as required. Potentially in some instances,reliability could be so high and production cost so low, that
the modules could be thrown away upon failure. Capabilities
of a certification facility should be considered when deciding
on a depot repair facility. Surge capability in time of conflict
must also be considered in the decision. As the functional
complexity of the line replaceable element is reduced, the
4 2,
.9
available industrial base capable of performing the repair
expands. LRMs for example would be more appropriate for con-
tracted depot repair than LRUs. Standardization of the elements
will expand the manufacturing base and the potential contract
maintenance facilities thus improving surge capability in time
of conflict.
4.4.4 Maintenance Strategy
Maintenance alternatives that may be considered for
use with modular avionics include the scheduling of preventa-
tive maintenance versus the use of maintenance on demand, as
traditionally practiced with aircraft avionics suites. Figure
4.4-2 shows the alternatives being considered.
A-32405
SCHEDULEDINTERVALS
MAINTENANCEFREQUENCY
ON INDICATIONOF FAILURE
- Figure 4.4-2 Maintenance Strategy
The decision criteria to be considered in selecting
the alternatives include:
1. Mission reliability
2. Hardware reliability
3. Operation and support cost.
As the predictive capability of OBT improves and theunderstanding of avionics hardware failure mechanisms increases,
4-25
0""
it may be possible to identify expected intervals between fatigue
failures in avionics hardware. It will be possible to set
maintenance intervals shorter than the expected failure free
interval so that the line replaceable element can be removed,
inspected, and recertified for a new failure-free interval.
Such use of scheduled maintenance could significantly improve
mission reliability and system availability. Operation and
supp,'rt costs may be reduced by reducing unscheduled mainte-
i.aice demands. Scheduled maintenance may require environmental
data from the TSMD to provide the detailed knowledge of environ-
mental exposure necessary for failure predictions.
a4* 4 -26
5. SCENARIO DEVELOPMENT
It is necessary to make assumptions regarding systems
characteristics and an operating environment in order to use
simulation models to conduct a cost/benefit study. This oper-
* ating environment and the system description form the scenario
in which the study is conducted. In planning studies such as
Air Force 2000, the support requirements are predicted to
4 change in order to adapt to a much more dynamic battlefieldenvironment. Mobility, flexibility, and survivability must be
built in to the support system. This forces decentralization
of the support structure with more reliance on self-contained
diagnostics including detection and isolation of faults. There
will be fewer main support bases and they will be located well
out of the combat area. .The main support bases will supplyoperating bases from which aircraft will deploy to dispersed
operating locations. Autonomous operations from these loca-
tions would be expected without relying on the support equip-
ment typical of present fighter aircraft. A modular avionics
architecture could be utilized on fighters, transports or
bombers in the planned scenario, but the worst case scenario
would be that of the fighter aircraft. In most planning sce-
narios, the transports and bombers would not be operating from
the austere dispersed operating locations that would be ex-pected of the fighters. The fighters generally have more rig-
orous avionics requirements brought about by size limitationsdriven by performance and radar cross section requirements.
Single engine propulsion systems make control electronicsflight critical and thermal management is more difficult to
achieve when cooling capability is carefully budgeted. The
offensive and defensive fighter avionics present a formidable
a 5-1
challenge to the integrator who must fit the avionics hardware
into the available space. Present fighter aircraft will serve
as the most appropriate information base to use in the simula-
tion. Today the typical fighter aircraft completes approxi-
mately 250 sorties per year. Each sortie lasts approximately
1.5 flight hours. As noted in the background information, the
aircraft of the 1990s will be expected to have a sortie surge
rate of up to five per day and each of the sorties could last
as long as four hours.
5.1 AVIONICS ARCHITECTURE BASELINE
The avionics architecture for mid 1990s application
has not yet been defined, but research into some of the de-
sirable characteristics of the architecture for tactical air-
craft has been underway for some time. The avionics suites of
aircraft like the F-15E or F-16C/D aircraft offer the most
substantial information applicable to mid-1990s fighter re-
quirements. Current laboratory development work on systems
such as the ICNIA, URR, CSP, and the VHSIC MIL-STD-1750A
Processor (VI750A) serve as the partitioning reference for
much of the module estimation.
In order to formulate a baseline for the study, the
general architecture shown in Fig. 5.1-1 has been assumed.
Although it does not represent the architecture of any air-
craft, it will serve as the basis for development of line
replaceable element counts for cost estimating. LRUs or LRMs
could be used as the hardware building blocks to fabricate the
*' avionics suite. For purposes of the study, the module types
5-2
A-24350
SUSYTE OUR DISPLAY
-- -- -- -
6RXI u
P= oMISIO
4DAT
REORE
S4AY.I
FigureN 5.1-1IC Modular Avinis ArchiecturCONROL
MISO 5-3A'p.S.It oa!111IL
~ y.' ~ .' :. * U - U U .~U ~ *U~UY4
4UM US .yf i
to be used as the basic elements are shown in Table 5.1-1.
" The LRUs are then formed from combinations of the LRMs and
packaged in a MIL-STD-1788 standard package. The use of threeintegrated racks is assumed for the suite of LRMs. The tasksperformed by the avionics suite, based upon current tactical
fighter requirements, are shown in Table 5.1-2.
TABLE 5.1-1
BASELINE COMMON MODULE SET
1. 1750A DIGITAL PROCESSOR/CSP ESU* 2. AVIONICS BUS INTERFACE
3. SHARED MEMORY MODULE/CSP GLOBAL MEMORY. 4. OPTICAL NETWORK INTERFACE/SENSOR INTERFACE
* 5. VIDEO MODULE/CSP VIDEO INTERFACE6. COMPLEX VECTOR PROCESSOR MODULE
7. FLOATING POINT PROCESSOR ELEMENT
8. MASS MEMORY
9. MIL-STD-1553 I/O MODULE10. IEEE 488 I/O MODULE/USER GUIDE CONSOLE INTERFACE11. MIL-STD-1760 STORES I/O MODULE12. DEDICATED INTERFACE MODULE
- ANALOG/DIGITAL
- DIGITAL/ANALOG@ - DISCRETES
- SERIAL CHANNEL (IN AND OUT)
13. CSP FLOATING POINT PROCESSOR ELEMENT14. CSP SORT ENHANCED PROCESSOR ELEMENT
15. CSP DATA NETWORK16. VECTOR PREPROCESSOR
17. IMAGE PROCESSOR18. BI-PHASE CORRELATOR
* 5-4
W *- - - .. -- "L... -." , " -* ,
TABLE 5.1-2
AVIONICS TASKS
Communication Navigation Identification
UHF Comm IFF Transponder Intercom
VHF Comm IFF Interrogator TACAN
* ILS/Flight Director INS
Vehicle Management
Flight Control Engine Inlet Scheduler
Air Data Engine Control
Radar
Antenna Signal Processor
Receiver/Transmitter Computer
Electronic Warfare
Dispensers Electronic Warfare Warning
Radar Warning Tactical Jamming
Mission Area
'. Fire Control Stores Management
Cockpit Video Electro Optical'P
When this task grouping is translated into modules
required, the quantities shown in Table 5.1-3 result. For the
LRU analysis, the modules should be repackaged into an appro-
priate number of LRUs, each containing an avionics bus inter-
face module and a power supply module. An initial partitioning
estimate for the number of LRUs to be included in the baseline
study is 60. This may be varied as more data on the specific
suite to be simulated is collected during the study effort.*.1
1e..
5-5
........ ... . . r " " . . . . .h.
V
TABLE 5.1-3
REQUIRED AVIONICS MODULE COMPLEMENT
SENSORS/ INTEGRATED RACK #1 INTEGRATED RACK 02 INTEGRATED RACK 03
1. !CNIA RF #1 Avionics Power Supply Avionics Power Supply Avionics Power Supply
2. TCNIA RF #2 Optical Network Optical Network Optical Network
3. INS #1 Avionics Bus Interface Avionics Bus Avionics Bus
4. INS #2 Optical Network Optical Network Optical Network
5. Electronic Control Avionics Bus Interface Avionics Bus Interface Avionics Bus Interface
6. Pneumatic Sensor Optical Network Optical Network Optical Network
7. Radar Antenna Video Module Video Module Video Module
- 8. Low Power RF Floating Point PE (8) Floating Point PE (2) Floating Point PE (2)
S9. Radar Rec/XTR 1750A DP (2) Sort Enhanced P1 (4) V1750A DP (2)
10. Chaff Dispenser Element Supervisor (8) 1750A DP (2) Element Supervisor (4)
- 11. EW Receiver Global Memory (4) Element Supervisor (6) Global Memory (2)
12. EW AMP Data Network (8) Global Memory (2) Data Network (4)
IJ. EW AMP Avionics Bus Interface Data Network (8) Avionics Bus Interface
14. EW AMP 1750A DP Avionics Bus Interface NV 1750A
15. RIU - Stores Avionics Bus Interface NV 1750A DP Optical Network Interface
16. RIU - Stores Avionics Bus Interface Avionics Bus Interface Mass Memoryrp
17. RIU - Stores Avionics Bus Interface Avionics Bus Interface
18. RIU - Stores Avionics Bus Interface NV 1750A DP
19. HUD NV 17SOA DF Avionics Bus Interface
20. Display Avionics Bus Interface Avionics Bus Interface
21. Display Avionics Bus Interface Avionics Bus Interface
22. Display VMS Power Supply NV 1750A DP
23. Control Avionics Bus Interface Aviori -s Bus Interface
24. Control Avionics Bus Interface
25. Control MS Power Supply
26. CVTS Avionics Bus Interface
27. CVTS
28. Intercom
28 LR/195 SRU 41 LRMs 41 LRMs 35 LRMs
.4
5-6
Ire
5.2 ELEMENT DESCRIPTION
The module count is sensitive to size allowances. A
5.88 x 6.4 inch card size is used as the baseline. Multiple
printed wiring boards (P~)with varying pitch (or thickness)
requirements are permitted as a module. The line replaceable
elements are assumed to be protected from electrostatic dis-
charge (ESD) , electrical overstress (EOS) , electromagnetic
interference (EMI), and the handling environment. Conduction
cooling to the siderails is used in the baseline module. Itis anticipated that size may have a significant influence on
life cycle cost so element size may be varied during the sen-
0 sitivity analysis to measure the influence on life cycle cost.
IBM has projected the relationship between the total number of
modules required, and the size of the module, in work done in
conjunction with their CSP contract. This relationship, shown
in Fig. 5.2-1, may be used in the sensitivity analysis.
5.3 OPERATIONS AND LOGISTICS SCENARIO
The operational environment includes the number of
bases, their location, and the number of aircraft supported at
each base. Many models that might be used in the analysis
consider the base location in two categories:
(1) Within the continental U.S., (CONUS)
(2) Overseas.
The transportation costs for pipeline spares is
simulated as a fixed amount for each category. The aircraft
to be considered in the scenario will enter the inventory
5-7
A 32404
191200 IL
~1000
2 7975-
0 60LL0
Ij 40 40
~25
B C D E FULL AIRMODULE SIZE (STANDARD ELECTRONIC MODULE)
Figure 5.2-1 Number of Modules Required Versus Module Size
according to the acquisition and production schedule shown in
ft..Fig. 5.3-1. The acquisition schedule assumes a developmentphase for the preferred modular avionics concept which spans
6 approximately two years before production planning for the
first deliveries begins. For the scenario, the F-16 basing
scheme has been adopted. The simulation should use 8 CONUS
a, wings and 6 overseas wings. Bomber and transport quantities
.4 will be assumed to build at approximately the same rate as is
currently being maintained.
.5-
*5-8
-A
NL m
4 0 4
V/100
U. a.
z L m (A 0 (
- -L0 4 c 4 - i __ _ _ _ _ _
0- 0 - c c c c i< c
0 2 _ __WmW_ _ _ __(A_ _
CL__ - ~
3j 0
5-9-
The system support environment and the structure of
each maintenance level form the logistics support scenario.
Since the decision criteria for the partitioning option in-
cludes the ability to eliminate the intermediate maintenance
level, both two-level and three-level maintenance options
should be included. Figure 5.3-2 shows a simplified three-level maintenance model where LRUs are repaired by replacing
SRUs at the intermediate maintenance level (I-level). The use
of an AIS will be assumed for this level. SRU and LRU spares
are stocked at I-level. SRUs removed at I-level are sent back
to depot for repair or condemnation. The long pipeline be-
tween I-level and depot is filled with SRUs and the sparing
0policy focuses on SRUs.
A 324M
" O-LEVEL LRUs REMOVED 1
~AND REPLACED
LRUs VERIFIED
SPRE SRUS REMOVED.-- AND REPLACED
"" -DPOT ESRU$":':, U" _[O REPAIRED
SPRE (ORJ CONDEMNED)
" -. Figure 5.3-2 Simplified 3-Level Maintenance Model 1
A two-level maintenance model is shown in Fig. 5.3-3.
"-...Both LRU and LRM spares are stocked at the organizational'-."level (O-level). There is no AIS in this arrangement. The
,," long pipeline between O-level and depot is now populated with
-0,either LRUs or LRMs. The sparing policy focuses on LRMs. The
e.'..5-10
"""
W w w - ' - .-u-- U r rr ' w -r r' . n W W W,=t Wfi ~ -. '% -e7 In -. ; - - '. - -; ;'' ",V' T w 7 r u -v , 'n wV r- r r
support cost in either of the maintenance models is dependent
upon the number and cost of the elements flowing through the
long pipeline.
O-LEVEL A 1
LRUiLRM REMOVED dLRU/LRMAND REPLACED SPARES
DEPOT
SRU/SRM UNIT REPAIRED OR CONDEMNED1 LRU/LRMSPARES SPARES
REMOVE AND REPLACE
SRU/SRM
REPAIREDIOR CONDE. MNF
Figure 5.3-3 Simplified 2-Level Maintenance Model
Important support cost drivers in addition to spares
cost are direct labor charges for repair. These charges include
replacement activities at each maintenance level and capital
investments for the AIS, AIS support equipment, AIS spares,
and AIS facilities. The AIS facilities include shelters, gene-
rators, tools, and other associated items. Since direct labor
is a function of unit reliability, it can be estimated when
the reliability has been defined.
For the cost analysis, each of the options should be
evaluated using both the two level and the three level models.
A realistic implementation of line replaceable elements may
require use of both two level and three level maintenance.
High failure, high cost elements would be maintained using
three levels while all remaining elements would be maintained
using the two level concept. In this case, a reduced AIS capa-
bility would be retained in the field but the cost estimate
would be reduced to reflect the reduced effort at the intermed-
*" iate level.
5-qN 5.-6d
I,
6. ANALYSIS METHODOLOGY
This cost/benefit study has been formulated to aid
2decision makers in selecting a preferred modular avionics arch-
itecture concept. The factors influencing the formulation of
the concept have been discussed in Chapter 3. These factors
include optional features that must be evaluated using the
decision criteria developed in Chapter 4. The impact on the
* life cycle cost of the system must be considered for each of
the alternative decisions. Because of the interactions among
features it is not always possible to develop an independent
cost estimate for one feature alone. For example, the charac-
teristics of the OBT utilized may influence the results of thelife cycle cost analysis when the Integrated Maintenance
Information System is implemented. The life cycle cost of the
system incorporating both features may be less than the cost
of either feature evaluated separately. However, some factors
will be able to be treated independently. The certificationfacility, for example, may not directly influence element
performance or reliability in any quantifiable manner. It
will only reduce the risk of .introducing non-compliant ele-
ments into the inventory. The certification facility may be
- treated as an independent factor for which the investment cost
necessary to implement and maintain it would be estimated.
It will be necessary to develop life cycle cost esti-
mates for all combinations of the dependent factors unless
further simplifying assumptions are made. It may be possible
to select the primary alternatives in sequence without consid-
ering the influence of all associated 'factors. This strategy
. ..
4.%8
would require that the analyst (14 -e r I ,oT: interactions
amen< factors that wou,_ld ,he expecte t. be minor in order to
reduce the combinations of factors that muist be evaluated to
determine the preferred i, :'u-ar a"ionmc c-ncopt.
To make intel-igent acquisit11(. Vetisions regarding
the selection c: modular ar ionics it is recessary to look
beyond the immediate cost o7 cevelcping and producing the
avionics. What may appear to be an expensive alternative
among competing architecture concepts may be the least expen-
sive when operation and support costs are considered. For
this reason, it is necessary to consider the life cycle cost
when making decisions. It should be noted tat LCC is not the
same as Operating and Support (O&S) cost, but rather includes
O&S as an element of cost. The modular a,'ionics LCC has been
captured when the estimate includes the cost to develop, pro-
duce, operate, and support it. Figure 6-1 illustrates the LCC
equation. LCC elements foil,, the modular avionics life cycle
and often overlap as Figure 6-' demonstrates.
DEVELOPMENT COST + PRODLC iON CoST =URQAM ACQUI S1: Lot COST
OPEPATING COST + SUPPORT COST OWNERSHIP COST
PROGRAM ACQUISITION COST + OWERSHIP COST" LIFE CYCLE COST
Figure 6-1 Life Cycle Cost Equation
Development costs include. all costs required to develop
* the modular avionics nr~or to a ccm.r-itment "or production.
This includes the cost o- en2eeorinrc design, the manufacture
or test sets, and testing , prove l.es :4 gn. The development
costs include al cr, rractual and Governr-nent co.s ts incurred to
* conduct and support the following:
".
* .-- -... ................................................................. '.-.- .....-..-"...--".-' ....."-"-".'".?,'- -,K '-'".. ~~~ ~ ~~~~~~~~~~~~ ~~~~... .............. . .. . .". " . . ... - . " .%- .. ..- .- ,- " %, t"
* Basic and Applied Research
* Engineering Studies and Analysis
0 Exploratory and Advanced Development
Selection of materials, components,processes
Engineering, design, fabrication,manufacture, and test of engineer-ing models of system componentsand related support
. Full Scale Development
Engineering, design, fabrication,* manufacture and test of development
articles and related support
Government test and contractor sup-port
Software development
]UFLS-CYCUE COST
. ACQUISTION COST
• 6-3
0 4-OW VNHI COST
Icauw~ IOPRATNIS ~AND
POOUCTICOST
IONIA O. W I TW
AM GRM P
Figure 6-2 Modular Avionics Life Cycle Flow
* 6-3
IRw7-W ER R -w wvri~vr
Production costs include the costs associated with
the fabrication, assembly, and delivery of the modular
avionics. Obviously the number and complexity of components
for a particular alternative will be a major production cost
driver. Production costs are, however, a minor portion of the
total LCC of modular avionics.
The most important costs in the modular avionics life
cycle are the ownership (operating and support) costs. Although
some avionics components may not have a 20 year lifetime assumed
for LCC analysis, due to technological improvement or future
risk factors, the ownership costs must be determined on a stand-
ard basis. These costs start with the modular avionics delivery
and continue throughout the operational life. The ownership
costs are summarized as:
* Operational personnel cost
* Maintenance personnel cost (differingfor LRU or LRN approach)
M ~aintenance material and facilities cost
* * Ground support equipment cost
* Base operating support cost
* Transportation (pipeline) cost
0 Personnel training cost
The computation of LGG requires:
(1) Hardware configuration description
4(2) Operations and logistics scenario
(3) Defined system life cycle
(4) Associated development and investment
activities.
6-4
0 Basic and Applied Research
* Engineering Studies and Analysis
* Exploratory and Advanced Development
Selection of materials, components,processes
Engineering, design, fabrication,manufacture, and test of engineer-ing models of system componentsand related support
* Full Scale Development
Engineering, design, fabrication,4 manufacture and test of development
articles and related support
Government test and contractor sup-port
Software development
UPE-CYCtS COST
P "O- ACOUSN coS
0 4.-O W1111S COST--------0
c
"co
COUMWY'JAL IOPENAIU
YAM
vOLA- P_ SURPOT
I -
Figure 6-2 Modular Avionics Life Cycle Flow
4 6-3
Production costs include the costs associated with
the fabrication, assembly, and delivery of the modular
avionics. Obviously the number and complexity of components
for a particular alternative will be a major production cost
driver. Production costs are, however, a minor portion of the
total LCC of modular avionics.
The most important costs in the modular avionics life
cycle are the ownership (operating and support) costs. Although
some avionics components may not have a 20 year lifetime assumed
for LCC analysis, due to technological improvement or future
* risk factors, the ownership costs must be determined on a stand-
ard basis. These costs start with the modular avionics delivery
and continue throughout the operational life. The ownership
costs are summarized as:
0 Operational personnel cost
0 Maintenance personnel cost (differingfor LRU or LRM approach)
0 Maintenance material and facilities cost
& Ground support equipment cost
* Base operating support cost
0 Transportation (pipeline) cost
0 Personnel training cost
The computation of LCC requires:
*(1) Hardware configuration description
(2I) Operations and logistics scenario
(.3) Defined system life cycle
(4) Associated development and investmentactivities.
* 6-4
The hardware description has been outlined in the
scenario development section, but additional definition may be
required to tailor the information to the analysis model being
used.
The operations and logistics scenario was established
in the last chapter, but it may be modified as required during
the analysis of alternatives.
A period of 20 years for the avionics life cycle is
appropriate for analysis, although technological advances and
* changes in the nature of the threat can sometimes be expected
to force avionics changes in a shorter time period.
'... The program acquisition (development) activities are
generally not modeled in most LCC models so it is necessary to
develop estimates off-line for input of program acquisition
costs to the LCC models. Where it is necessary to do so, the
cost elements should be estimated using the general process
outlined in Fig. 6-3. In practice a combination of three
methods is used:
(1) Bottom-Up Approach
* Each cost element in the process (i.e.,development engineering, printed wiringboard fabrication, etc.) is estimatedand the aggregate sum represents thecost.
0. (2) Analogy
The known cost of a similar system isA. modified appropriately to account for
differences and the result serves as anestimate of the analyzed system cost.
(3) Parametric Analysis
An analysis based upon a minimum numberof characterizations (such as size,
* 6-5
0 4c0 w0 w
00 W
4 4W X
a. a.000 00 0000
0 >.
0~~ X.aJ 4-1 .. Z
5 Lo
0
5c 5 4.Ui > -s 5 l
z Z z 0. z O4wO
<
w
cp 0I p .It~
'. N
weight, lines of code, etc.) to esti-mate cost based upon a large storeddata base. RCA Price Models utilizethis type of analysis.
Since it will be necessary to make determinations of
the system LCC for each combination of factors, a flexible
Y simulation model is most appropriate. The models used should
accommodate the introduction of a product performance agreement
and should be able to simulate the effect that standardization
of hardware across weapon systems has on LCC. A review of the
decision criteria established for the alternatives shows that
insight into the ownership (operating and support) costs will
* be important for decision makers. Typical detailed ownership
costs include:
Spares acquisition
Base level
Depot
Condemna tion
Support equipment acquisition
Flight line level maintenance cost
Base intermediate level maintenance cost
Depot level maintenance cost
Training* Technical data acquisition
Item management
Data management
Packaging and shipping
* Support equipment maintenance
Transportation
Shop spare
Inventory storage
Deactivation costs
Flexibility will be necessary in order to model the effect ofeach combination of alternatives in the simulation.
* 6-7
7. CONCLUSIONS
This study was commissioned to formulate an approach
to evaluate alternative features of a modular avionics archi-
tecture concept for the mid-1990s. Review of background infor-
mation shows that the operational support environment will be
characterized as mobile, flexible, and austere. The support
of avionics in the next decade must be more autonomous with
less reliance on well equipped secure support bases close to
the battlefield. The avionics rer, ired on the aircraft con-
- tinues to increase while the available space shrinks in order
to reduce radar cross section and to increase maneuverability.
Cost of avionics acquisition and support must be controlled as
the role of avionics expands and reliance on external support
.4 is reduced. The criteria by which each alternative feature is
measured includes, therefore, the impact on system life cycle
cost and the contribution to readiness or mission reliability.
The alternatives to be considered in a cost/benefit
study have been identified and the decision criteria to be
used in selecting them for inclusion in the preferred avionics
concept have been developed. It has been necessary to establish
a baseline scenario using available information on the aircraft
and avionics expected to be used in the next decade. The
research work on advanced avionics architectures using VHSIC
technology sponsored by the laboratories at Wright-Patterson
AFB has been used to construct an avionics suite for use in
the study. The baseline architecture is described for a
fighter aircraft since it represents a worst case application
of the modular avionics concept. Phase I VHSIC with 1.25
7-1
micron feature size is used where possible in the module designs
since that is the technology used in the design process today
that will be in the field in the mid-1990s.
Interaction among the factors will make it difficult
to establish a modular avionics program unless a careful eval-
uation is made of combined factor effects. For example, the
use of effective OBT coupled with reliability improvements in
the line replaceable elements may make it possible to eliminate
the intermediate maintenance shop. With the intermediate shop
gone, the benefits of programs such as MATE and IMIS are ex-
pected to be reduced. The costs must, therefore, be developed
for each combination of factors to be considered.
The analysis methodology to be used in the study has
been discussed with regard to the desirable characteristics of
a simulation model to be used in developing the life cycle
costs. These costs can be used in completion of evaluation
tables similar to that shown in Table 7-1.
7-2
'4A
-'E-4
4044
U7-3
MII'aiulI OW
4I REFERENCES
1. Dougherty, "Fighter Avionics Supportability Demonstra-tion," undated briefing material.
-~R-1
BIBLIOGRAPHY
"Module Development and Implementation Plan," Directorate of
Avionics Standardization and Systems Architecture, WPAFB, OH,
September 1975.
"Future Fighter Avionics Demonstration Program," General
Dynamics Corp., Report No. 16PR3726, 15 May 1985.
"Life Cycle Cost Comparison of Advanced Avionics Systems,"
TASC, Technical Information Memorandum 5043-1-1, 31 May 1985.
"An Integrated Fault Tolerant Avionics System Concept for
Advanced Aircraft," Report No. R-1226, Charles Stark Draper
Laboratories, Cambridge, MA, Februarv 1979.
"On Board Test (OBT) Improvement Project - Phase II," Director-
ate of Support Systems Engineering, ASD-TR-83-5012, September
1983.
"Minutes of the First Open Forum on Modular Avionics Standards,"
ARINC Research Corp., Annapolis, MD, 26-28 June 1984.
Dougherty, "Fighter Avionics Supportability Demonstration."
Engelland, J.D., "Avionics Standardization From a Designer'sPerspective," IEEE/AEss Symposium on Standardization in Mili-
9 tary Avionics Systems Architecture, Dayton, OH, 28 November
1979.
9-.
f. Scott, D.M. and Mulvaney, S.P., "Common Modular Avionics:
Partitioning and Design Philosophy," Proceedings of the AIAA/
IEEE Sixth Digital Avionics Systems Conference, Baltimore, MD,
3-6 Dec 1984.
B-I
Radcliffe, W. and Boaz, J., "Modular Standards For Emerging
Avionics Technologies," Proceedings of the AIAA/IEEE Sixth
Digital Avionics Systems Conference, Baltimore, MD, 3-6 Dec
1984.
"-" Poradish, F., "High Density Modular Avionics Packaging,"
-Proceedings of the AIAA/IEEE Sixth Digital Avionics Systems
Conference, Baltimore, MD, 3-6 Dec 1984.
Austin, M. and McNichols, J., "Modular Avionics Packaging
Standardization," Proceedings of the AIAA/IEEE Sixth Digital
Avionics Systems Conference, Baltimore, MD, 3-6 December 1984.
"Standard Electronic Module Radar Life Cost Comparison,", Battelle Columbus Laboratory, AFAL-TR-79-1025, April 1979.
"Standard Electronic Module (SEM) Implementation Study,"
Report No. P78-84, Hughes Aircraft Company, Culver City, CA,
June 1978.
w.
4%.
S.
SUS GOVERNMENT PRINTING OFFICG 1988 ,48 054 H 19
B-2
J
J
.1
3
'4
*
I
).1
I /'U,
et 6 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
- 'U. ~